Over the past decade, e-commerce has transformed the global retail landscape, reshaping how goods are purchased, distributed and delivered. At the centre of this transformation lies last-mile delivery (LMD), the final and most consumer-facing stage of the logistics process, which directly influences service quality and customer satisfaction (Deloitte 2020). However, ensuring efficient and reliable LMD remains a persistent challenge for retailers and logistics providers, particularly in regions where infrastructural, regulatory or residential factors restrict access to consumers (Samuels, Takawira & Mbhele 2024). These challenges are evident especially in urban areas, where population density, safety concerns and delivery access restrictions can hinder smooth operations. The global adoption of online shopping continues to accelerate, with a growing share of consumers worldwide embracing e-commerce platforms.

In developing countries such as Malaysia, approximately 80% of the population engages in online shopping (Naseri et al. 2021), indicating the rapid expansion of digital retail channels. By contrast, developed countries in Europe reported e-commerce turnover grew more moderately by 6% in 2022, reaching approximately €899 billion, as inflation and reduced consumer spending slowed overall growth despite continued expansion in online service sales (Ecommerce Europe 2023). Developed countries benefit from advanced logistics, widespread internet access and mature digital payment systems, while developing countries continue to face unique challenges related to delivery reach, digital literacy and infrastructural limitations (Banga & Banga 2022). These disparities highlight the importance of considering regional development levels when examining online shopping trends.

In South Africa, the e-commerce sector expanded by 66% between 2018 and 2021, growing from R14.1 billion to approximately R40 billion and accounting for 4% of all retail in the country (Business Insider SA 2021). Takealot, a leading South African online retailer, reported 32% year-on-year revenue growth to US$385 million (approximately R6.7 billion), reflecting its strategic operations, expansion of subscription services, and responsiveness to evolving consumer demands (Ndlovu 2025). The global growth of e-commerce has been driven by increased Internet and smartphone penetration, social media influence, digital marketing, globalisation and enhanced logistics networks (Gupta et al. 2023). These technological and social drivers have significantly influenced the purchasing behaviour of young people, shaping their decisions regarding product selection, timing and purchase channels (Chopra, Gupta & Manek 2020; Farah et al. 2018).

Farah et al. (2018) investigated internet usage and online shopping, revealing that university students aged 18–30 are among the most active internet users and online shoppers, representing a key segment of young consumers in the digital marketplace. Similarly, Chopra et al. (2020) emphasised that this group increasingly relies on online channels for product discovery, peer reviews and price comparisons, shaping how and when they shop. Convenience, affordability and access to promotions are key motivators, particularly for students navigating demanding academic schedules and limited budgets (Al Asheq et al. 2022; Lim et al. 2024; Tian & Zhang 2024). For the purpose of this study, the term student refers to a young consumer aged between 18 and 30 years, who represents the primary user base for online delivery services.

As consumers shift towards online shopping for convenience and time-saving benefits, traditional brick-and-mortar stores have experienced a corresponding decline in sales (Moeti, Mokwena & Malebana 2021). In this evolving landscape of online shopping, customer satisfaction is influenced by factors such as convenience, accurate real-time information, data security, competitive pricing, responsive support services and reliable delivery (Vakulenko et al. 2019). Fast and timely deliveries are essential, with many customers preferring same day or next day delivery options (Olsson, Hellström & Vakulenko 2022). Companies that strategically optimise LMD operations often report enhanced customer loyalty, higher sales and improved profitability (Puri 2022).

Last-mile delivery refers to the final stage of the transport process, encompassing the movement of goods from a fulfilment centre or retail establishment to customer’s chosen location, typically their home or office (Deloitte 2020). Consumers expect fast and flexible delivery options, accompanied by features such as real-time tracking and transparent communication (Tsao & Tseng 2021). Diverse delivery options such as home delivery, click-and-collect, locker points and express delivery have been shown to enhance customer engagement (Milioti, Pramatari & Kelepouri 2020). According to DispatchTrack (2020), a global provider of real-time delivery management solutions, 51% of consumers avoid e-retailers with limited delivery options, whereas 96% remain loyal to e-retailers that provide a positive delivery experience. These findings highlight the growing importance of efficient and customer-centric delivery processes in shaping online shopping behaviour globally.

In South Africa, where e-commerce continues to expand, similar trends are observed, with consumers expecting flexible, reliable and transparent delivery options (Samuels et al. 2024). Tailored and efficient LMD services significantly boost customer retention and overall satisfaction (Puri 2022; Wang, Zhang & Huang 2021). The LMD process is therefore essential in shaping customer experience and influencing satisfaction levels (Pourmohammadreza, Jokar & Van Woensel 2025). However, in certain contexts, particularly student accommodation, which formed the focus of this research, LMD is often constrained by residential restrictions. Many university campuses and student residences enforce strict policies that limit courier access for safety reasons, preventing vehicles from entering campuses or staff from accepting parcels on behalf of residents.

Similar restrictions have been reported internationally. For instance, Cherrett et al. (2017) found comparable restrictions at the University of Southampton in the United Kingdom (UK), as did the University of Bristol (2023) in the UK and Gao (2022) at the military school campus in China. Students often face various challenges in receiving parcels, frequently having to collect them outside the campus grounds, sometimes far from their accommodation, causing significant inconvenience. In China, studies indicate that approximately 25.84% of students are required to collect parcels off campus, often encountering long queues and congestion at distribution points, especially after class hours (Gao 2022; Xiang & Wu 2018). These delivery challenges are applicable to a South African context that compromises the convenience and reliability of the LMD experience for students, causing dissatisfaction and prompting behavioural shifts, such as opting for alternative delivery services or abandoning online purchases altogether.

Given students’ growing role in the online consumer shopping base, these delivery challenges warrant urgent attention. Deloitte (2021) reported that 13% of online shoppers in South Africa are aged 18–24, while 31% are aged 25–34. Similarly, Cowling (2024) found that 18–24-year-olds account for 30.9% of social media users, and those aged 25–34 years represent 30.7%. These figures highlight the increasing digital engagement and purchasing potential of young consumers. Hence, efficient delivery solutions and responsive customer service are crucial for e-retailers seeking to improve satisfaction and build long-term loyalty within these young consumers. While there is a growing body of literature on online consumer behaviour, service quality and repurchase intentions (Pentz, Du Preez & Swiegers 2020), relatively fewer studies have explored the specific logistical challenges of LMD experienced by students.

Internationally, researchers have explored the impact of logistics on student online shopping experiences (Cherrett et al. 2017; Choi 2019; Jalil 2018; Mahdi Zarei, Chaparro-Pelaez & Agudo-Peregrina 2020), yet within the South African context, it remains limited. Locally, studies have primarily focused on general online shopping trends, consumer trust and delivery efficiency, without addressing the specific challenges students encounter when accessing deliveries from restricted accommodation settings. This gap provides the rationale for this study, highlighting the need to investigate how delivery constraints affect students’ online shopping experiences and behaviours. This study explores the impact of delivery restrictions on LMD experiences of university students residing in student accommodation in Johannesburg, specifically in Auckland Park and Braamfontein areas.

National data indicate that young consumers aged 15–34 in South Africa control an annual spending pool of approximately R303 billion, with students alone contributing an estimated R41 billion annually (BusinessTech 2014; Student Village & Youth Dynamix 2023). According to Gregory and Rogerson (2019), Auckland Park and Braamfontein are student-dense areas, hosting students from the University of Johannesburg, University of Witwatersrand and private institutions such as Milpark Education Business School, Rosebank College and the South African School of Motion Picture Medium and Live Performance (AFDA). Understanding students’ delivery experiences is essential, as they constitute a significant portion of the online shopping population and represent a growing market with future purchasing power (Dlamini & Chinje 2019; Heyns & Kilbourn 2022). Moreover, as Gen Z’s demographic continues to shape the direction of digital commerce, the importance of adapting delivery strategies to meet their specific needs becomes evident.

Therefore, this study explored the impact of delivery restrictions on LMD among students living in student accommodations in Auckland Park and Braamfontein, Johannesburg. By focusing on this often overlooked yet growing segment of the e-commerce market, the study contributes to existing literature and provides insights for stakeholders in e-commerce industry regarding how delivery constraints influence students’ experiences, satisfaction and repurchase intentions. The article is structured as follows: a review of relevant literature, a presentation of the research methodology, a discussion of the findings and their implications, and finally practical recommendations and direction for future research.

This literature review examines the existing research on online consumer behaviour and LMD, with particular attention to the logistical challenges faced by students. Globally, studies have shown that delivery efficiency, service quality and flexible last-mile options significantly influence customer satisfaction and repurchase intentions (Aljohani 2024; Bin 2024; Olsson et al. 2022). In China, for example, approximately 25.84% of students are required to collect parcels off campus, often facing long queues and congestion at the distribution points (Gao 2022; Xiang & Wu 2018). Locally, research has primarily focused on general online shopping trends, consumer trust and delivery efficiency of young consumers (BusinessTech 2014; Dlamini & Chinje 2019; Heyns & Kilbourn 2022). While these studies provide valuable insights into the behaviour and expectations of young consumers, understanding the strategic role of LMD in e-commerce is critical for addressing the specific logistical challenges that affect students’ satisfaction and loyalty.

This study employed a descriptive survey research design to explore the impact of delivery restrictions on student accommodation in Auckland Park and Braamfontein, Johannesburg. The research instrument was developed using the electronic physical distribution service quality (e-PDSQ) framework and Expectation-Confirmation Theory (ECT), which are suitable for exploring last-mile challenges experienced by students in restricted accommodation settings. The instrument comprises originally developed items that captured the key dimensions of e-PDSQ such as reliability, responsiveness, flexibility, security and convenience. While ECT constructs are related to expectations, perceived performance and satisfaction. A quantitative approach was adopted, grounded in a positivist paradigm and deductive reasoning, to facilitate statistical analysis of structured data. The unit of analysis consisted of students residing in student accommodation in Auckland Park or Braamfontein who had made online purchases and had used LMD services at least three times in the past 12 months (2023–2024).

This criterion ensured participants had relevant and recent experiences with LMD services. The aim was to include students with diverse demographic backgrounds such as age, gender, academic level and cultural background to ensure comprehensive and representative findings. Participants were recruited across multiple and diverse student accommodations such as on-campus and private student accommodations to enhance representativeness and ensure relevance to the research objectives. Stratified and convenience sampling techniques were combined to balance representativeness and feasibility. Stratification ensured inclusion across diverse institutions and residence types in Auckland Park and Braamfontein, while convenience sampling facilitated access to respondents through digital platforms. The combination of both approaches supports the generalisability of the findings to the broader student population in similar urban settings, while noting the study’s focus on Johannesburg as a contextual boundary.

Cochran’ formula was used to determine the most appropriate target sample size, as the exact population of students in the two areas was unknown, and the target sample size was calculated as 384. A self-administered online questionnaire was distributed via WhatsApp, Instagram and LinkedIn. Validity was established through expert review and a pilot test of the questionnaire by distributing the online questionnaire to 10 students. Based on the feedback, minor modifications were made to improve comprehension. Reliability was assessed using Cronbach’s alpha, with coefficients above 0.7 for all dimensions, indicating strong internal consistency. Data were collected cross-sectionally between December 2024 and January 2025; a total of 474 valid responses were collected, and data were analysed using Statistical Package for the Social Sciences (SPSS), version 29.

Descriptive statistics summarised demographic variables such as gender, age, residential area, type of student accommodation, source of income and monthly expenditure on basic needs such as food, toiletries and transport. As demonstrated in Table 1, the majority of respondents were female (69.8%), while 29.3% identified as male, and 0.8% did not disclose their gender. This aligns with previous research, indicating that female participation is higher in the online survey (Becker 2022; Heyns & Kilbourn 2022). Research on LMD and repurchase intentions, online shopping perceptions and campus express delivery services have shown a consistent pattern of higher female response rate, ranging from 58% to 71.11% during the year, suggesting a higher female representation in studies related to e-commerce and delivery services (Chelvarayan, Jie & Fern 2021; Rajendran & Wahab 2022).

The age distribution of respondents indicates that 50.6% were between 22 and 25 years, while 45.6% were between the ages of 18 and 21. These findings align with previous studies indicating that individuals aged 18 to 24 are the most active online shoppers (Cherrett et al. 2017; Deloitte 2021) and reflect the typical student age range of 20–24 years in South African higher education (Department of Higher Education and Training [DHET] 2024). This suggests that the age distribution of respondents reflects the typical age profile of university students in South Africa. Regarding financial support, 49.8% of the respondents receive an allowance from parents or relatives, 44.9% receive bursary allowances, and 5.1% earn an income through employment. Monthly expenditure on essential items was predominantly between R500 and R1999, with 27.6% of the respondents indicating that they spend between R1000 and R1499 on essentials. This is followed by the respondents spending between R1500 and R1999 per month, while 19.0% spend between R500 and R999 monthly. Students’ online shopping behaviour is influenced by affordability and available income, which likely affect purchasing frequency and product selection. These financial patterns are important for understanding their online/parcel delivery preferences and constraints. The next section focuses on their experience with LMD, highlighting the delivery issues students commonly face. The following section explores students’ shopping behaviour.

The survey revealed that 62.4% of students live in residences where staff cannot receive parcels, leading to missed deliveries and, for 13.1%, missed academic commitments. This underscores the need for on-site solutions such as smart lockers or package rooms (Cherrett et al. 2017; Gao 2022), revised policies allowing staff acceptance with digital verification (Crystal Services 2024), and extended collection hours to evenings and weekends (Bouhouras et al. 2024). Satisfaction scores for arranging deliveries (M = 3.28), communication about delays (M = 3.24) and flexibility under access restrictions (M = 3.21) indicate service gaps. Couriers should introduce customised delivery windows aligned with student schedules (17:00–19:00), improve real-time communication and expand re-delivery options (Allen et al. 2017). Partnerships with universities to establish decentralised parcel hubs or micro-lockers in dense student areas such as Auckland Park and Braamfontein would reduce delays and reliance on peer handovers.

Over half of respondents (65.8%) avoided purchases because of unsatisfactory delivery options, citing high fees (39.4%) and long lead times (30.8%). E-retailers should respond with student discounts or free delivery thresholds, transparent pricing and simplified return policies to rebuild trust and encourage repeat purchases (Berthiaume 2023; Dablanc et al. 2017). Finally, 26.0% reported shopping less frequently and 25.3% used alternatives because of delivery restrictions, highlighting systemic gaps in student-oriented logistics. Policy incentives for shared lockers, community hubs or crowdshipping solutions (Das & Fianu 2018; Filiopoulou et al. 2022) could balance accessibility with security, while reducing students’ reliance on informal peer networks (56.5%).

The data collected for the study also showed that some students face academic trade-offs, such as missing classes or delaying study commitments because of restrictive delivery systems. This points to an underexplored consequence of last-mile inefficiencies: their potential effect on educational outcomes. Future research should therefore examine the link between delivery restrictions and academic performance, while also conducting comparative analyses of student housing delivery models internationally to identify scalable best practices. Further research of digital solutions such as locker apps, peer parcel-sharing platforms, and AI-driven delivery scheduling is recommended, as these could formalise the adaptive strategies already observed among students (Huang & Nuangjamnong 2023).

This study has shown that delivery restrictions in student accommodations significantly disrupt LMD, influencing both online shopping behaviour and satisfaction levels among university students in Johannesburg. Analysis of 474 survey responses revealed that prohibitions on staff receiving parcels, combined with high delivery fees and long waiting times, undermine the convenience and reliability that students expect from e-commerce. These constraints drive behavioural shifts such as reduced online shopping frequency, reliance on informal peer networks, and in some cases, academic disruptions where students miss classes to receive deliveries. While students demonstrate resilience by adopting coping strategies, including alternative delivery addresses, pick-up points and assistance from roommates, these adaptations expose broader inefficiencies in last-mile systems for student populations. The recommendations arising from this study call for student-centred delivery solutions within residences, improved flexibility and communication by logistics providers, affordability and transparency from e-retailers and supportive infrastructure and policy interventions from urban planners.

By connecting delivery inefficiencies to consumer trust and academic well-being, this research expands the discourse on LMD beyond operational performance to the lived experiences of student consumers. Theoretically, the study contributes by applying and extending the PDSQ and ECT frameworks to a unique student housing context, demonstrating their relevance for analysing both logistics performance and behavioural adaptation in constrained environments. Without targeted interventions, restrictive delivery practices risk eroding satisfaction and repurchase intentions among one of South Africa’s most digitally active demographics. Future studies should therefore explore the comparative effectiveness of international housing delivery models and assess the longer-term implications of delivery restrictions on both educational outcomes and consumer behaviour. The findings emphasise that LMD in student accommodation is not only a logistical challenge but also a determinant of digital inclusion, consumer confidence and academic engagement.